Cells of the innate immune system constantly evaluate host mucosal surfaces and peripheral tissues for signs of infection or injury. The host must find a balance between tolerance of beneficial microorganisms and minor non-pathological microbial encounter vs. the development of a robust immune response to more serious infections. Emerging evidence suggests that this decision is made by the cell based on the combinatorial signals it receives from its engagement with microorganisms and endogenous stimuli. These signals are sensed primarily by various classes of pattern recognition receptors (PRR), and while there has been remarkable progress in characterizing the individual signaling pathways induced through these receptors, relatively few studies have addressed how immune cells integrate combined PRR inputs and the combination of these signals with others arising from soluble host derived substances such as cytokines, lipids, and complement components. This project seeks to define the control principles that determine the complex relationship between signal input and output function in this scenario, and ultimately to generate quantitative computational models to describe cellular behavior in circumstances relevant to infectious disease. Using macrophages as a model system, we are initially characterizing the cellular response to PRR ligands and intact pathogens by measurement of a variety of readouts such as; signaling protein phosphorylation, intracellular trafficking, pathogen replication, transcription and production of immune mediators. We have profiled the response of macrophage cells to a group of toll-like receptor (TLR) ligands (LPS, Pam2CSK4, Pam3CSK4, Resiquimod 848 and Poly I:C). Analysis of the response to combined stimuli (mimicking what would occur with an intact pathogen) shows non-additive levels of activation of downstream signaling pathways. We have generated comparative transcriptional profiles following stimulation of RAW cells and BMDM with single or combined TLR ligands. These data suggest that the non-additivity in signaling outputs and cytokines is reflected at the transcriptional level. In ligand combinations that signal exclusively through the MyD88 adapter, all induced genes show less than additive responses. However, in ligand combinations including a TRIF-activating ligand, while the majority of transcripts remain less than additive, we see a select subset of genes induced to greater than additive levels. This selective outcome for combined MyD88+TRIF activation is likely used by the host as a detection mechanism for either simultaneous exposure to viral and bacterial pathogens, or to a significant infection with intracellular pathogen, and it leads to the increased production of a selective set cytokines that serve to drive a robust adaptive immune response. We seek to determine the molecular mechanisms underlying both the suppressed response to combined ligands activating common pathways and the synergistic responses to ligands that combine MyD88+TRIF activation. This year, to address how the TLR signaling network might mediate the suppressed response through combined MyD88 stimuli, we investigated the localization dynamics of proximal signaling components in the TLR pathway. We created a set of fluorescently tagged expression constructs to provide CFP and YFP fusions of the TLR receptors, signaling adaptors, and kinases involved in proximal signaling. Where possible, these fluorescent fusions were stably expressed in immortalized mouse macrophages (IMM) derived from mice deficient in the TLR signaling component in question. We stimulated cells with a matrix of single and combined TLR ligands and observed the effects on localization of pathway components. This study has identified signaling protein localization patterns that are differentially induced in cells challenged with single or combined ligands, and may indicate a basis for signaling attenuation upon combined TLR stimulation. This work is ongoing. We also seek to identify the basis of the greater than additive release of two key cytokines, IL-6 and IL-12, from macrophages in response to ligands which induce the TRIF and MyD88-dependent pathways. Identifying the cellular mechanism underlying this non-linear response will have important implications both for modeling of PRR pathway crosstalk in macrophages and also for identifying therapeutic targets for inflammatory disorders. We previously generated transcriptional profiles from BMDMs challenged with the TRIF-specific ligand poly I:C (I) and the MyD88-specific ligand R848 (R) and the combination (I+R) over a broad time course. This identified genes with expression characteristics that could implicate them as synergy factors underlying the enhanced production of IL6 and IL-12. This year, over two hundred genes identified by the transcriptional study were targeted in an siRNA screen to determine if they are required for the macrophage to induce high levels of these cytokines in response to the combined I+R stimulus. This screen has identified approximately 20 genes that appear to have important roles in regulating the response to combined MyD88+TRIF pathway activation. We are currently validating these hits and studying possible mechanisms of action. To evaluate the TLR signal integration that occurs in the context of a real infection, we previously initiated a study of the macrophage response to Burkholderia cenocepacia (Bcc), an opportunistic bacteria particularly problematic in cystic fibrosis and chronic granulomatous disease patients, and closely related to the category A select agents B. mallei and pseudomallei. Macrophages are likely to play a key role in Bcc-induced pulmonary infections, but very little is known about the mechanism of Bcc infection and replication in these cells. This year, we published a comprehensive characterization of the intracellular life cycle of B. cenocepacia and its interaction with the autophagy pathway in human macrophages (Al-Khodor et al. Cell Microbiol. 16(3):378-95). Electron and confocal microscopy analysis demonstrated that the invading bacteria interact transiently with the endocytic pathway before escaping to the cytosol. This escape triggers the selective autophagy pathway, and the recruitment of ubiquitin, the ubiquitin-binding adaptors p62 and NDP52 and the autophagosome membrane-associated protein LC3B, to the bacterial vicinity. However, despite recruitment of all the key autophagy initiation components, B. cenocepacia blocked autophagosome completion and replicated in the host cytosol. We found that a pre-infection increase in cellular autophagy flux can significantly inhibit B. cenocepacia replication and that lower autophagy flux in macrophages from immunocompromised CGD patients could contribute to increased B. cenocepacia susceptibility, identifying autophagy manipulation as a potential therapeutic approach to reduce bacterial burden in B. cenocepacia infections. In ongoing studies, we have initiated a study in collaboration with the LSB Cellular Networks Proteomics unit to use mass spectrometry to screen for candidate B. cenocepacia effector proteins that could mediate the autophagy subversion described above. Host autophagy pathway effectors will be affinity purified and analyzed for associated bacterial proteins. We are also developing an imaging assay to permit identification of host cell components that mediate the ubiquitination of cytosolic bacteria, which initiates the autophagy response. This assay will permit RNAi-based screening for components of the host sensing pathways for cytosolic bacteria.